Photo-realistic 3D building models reconstruted from RSI
Summary: Reconstructing the 3D building mesh from RSI based on DNN.
Past
The Design and Development of Open Geospatial Engine
Summary: Successfully directed a team of 20 highly skilled graduate students in the development of the Open Geospatial Engine, a cloud-native, large-scale remote sensing data analysis platform similar to Google Earth Engine. Upon its launch (Scheduled for December 2023), the OGE will offer tens of petabytes of Analysis-ready remote sensing data, provide online analysis services of hundreds of GIS operators and models, and enable online programming capabilities for users.
Multi-source 3D Information Management for 3D Spatial Data Infrastructure
Summary: Transitions of spatial data infrastructures (SDIs) support applications from 2D landscapes to 3D scenes. The existing methods for describing, managing, and providing services for 3D spatial data often lack coordination and efficiency. Moreover, the added complexity of 3D data structures necessitates novel approaches for component-level management and streaming capabilities. In response, we developed a generic conceptual model suitable for component-level management of diverse 3D spatial data in SDIs and discussed the design rationales and key considerations underlying the model. We formalized the flexible data composition and fine-grained lifecycle management in this model and specified this model at the cloud-optimized encoding level to enable efficient CRUD operations and streaming delivery of massive 3D spatial data. Our approach enabled direct streaming of the managed 3D spatial data without the need for redundant replication. We implemented, evaluated, and discussed the proposed approach in terms of service, accessibility, visualization, analysis cases, and efficiency. The results show that the proposed method is efficient in managing 3D spatial data and enables users to conduct 3D geo-analysis on the basis of specific parts of the data as needed. This work provides a scientific exploration that integrates the management and services of 3D spatial data in SDIs.
Unsupervised photo-realistic 3D Mesh segmentation based on machine learning
Outcome: A robust cloth simulation method was proposed for filtering ground primitives from airborne-based 3D meshes. Experiments demonstrate that the method can accurately decompose the realistic 3D meshes into the 3D ground model and non-ground 3D model, and it is a robust method suitable for different landscapes and is not impacted by vertex density and noise.
Outcome: I Proposed a novel bidirectionally greedy framework to extract fine-grained building models from photo-realistic 3D meshes in an unsupervised manner. The experimental results show that the accuracy and performance of the method even surpass that of supervised deep learning methods.
Virtual geographic environment platform for tailing dam failure simulation
Summary: Building upon the physical properties of tailings flow, a mathematical model for simulating tailings dam failures was constructed based on computational fluid dynamics (CFD), adhering to the principles of mass and momentum conservation. Based on the concept of VGE, a VGE platform for dam failure disaster evolution simulation is designed, which can simulate and predicts the route and disaster-causing range of tailings after a failure, and visualize the results in a three-dimensional, dynamic and interactive manner.
Contributed to the development of national/international standards
Outcome: Contributed to the development of the OGC international standards: OGC TrainingDML-AI
Outcome: Translate OGC CityGML Standard into Chinese
Outcome: Contributed to two Chinese national standards: Geographic Information, Digital Data and Metadata Preservation Part 1: Foundations and Geographic Information, Digital Data and Metadata Preservation Part 2: Content Specification for Earth Observation Data and Derived Products
OGEScript: An OGC-oriented Interoperable Script API for Online Geospatial Analysis
Code
PDF